drvi.internal.drvi.LatentStats

drvi.internal.drvi.LatentStats#

class drvi.internal.drvi.LatentStats(n_latent, vanished_threshold=0.5, dist_sync_on_step=False)[source]#

Track per-dimension min/max of the latent mean and report non-vanished dimension counts.

Observed bounds are accumulated over an epoch and promoted at reset() to serve as the reference bounds for the next epoch (also used by StreamingPairwiseMI for binning).

Attributes table#

`T_destination`
`call_super_init`
`device`	Return the device of the metric.
`dtype`	Return the default dtype of the metric.
`dump_patches`
`full_state_update`
`higher_is_better`
`is_differentiable`
`metric_state`	Get the current state of the metric.
`plot_legend_name`
`plot_lower_bound`
`plot_upper_bound`
`update_called`	Returns `True` if `update` or `forward` has been called initialization or last `reset`.
`update_count`	Get the number of times `update` and/or `forward` has been called since initialization or last `reset`.
`training`

Methods table#

`add_module`(name, module)	Add a child module to the current module.
`add_state`(name, default[, dist_reduce_fx, ...])	Add metric state variable.
`apply`(fn)	Apply `fn` recursively to every submodule (as returned by `.children()`) as well as self.
`bfloat16`()	Casts all floating point parameters and buffers to `bfloat16` datatype.
`buffers`([recurse])	Return an iterator over module buffers.
`children`()	Return an iterator over immediate children modules.
`clone`()	Make a copy of the metric.
`compile`(args, *kwargs)	Compile this Module's forward using `torch.compile()`.
`compute`()	Compute vanished statistics based on the previous epoch's bounds.
`cpu`()	Move all model parameters and buffers to the CPU.
`cuda`([device])	Move all model parameters and buffers to the GPU.
`double`()	Override default and prevent dtype casting.
`eval`()	Set the module in evaluation mode.
`extra_repr`()	Return the extra representation of the module.
`float`()	Override default and prevent dtype casting.
`forward`(args, *kwargs)	Aggregate and evaluate batch input directly.
`get_buffer`(target)	Return the buffer given by `target` if it exists, otherwise throw an error.
`get_extra_state`()	Return any extra state to include in the module's state_dict.
`get_latest_min_max`()
`get_parameter`(target)	Return the parameter given by `target` if it exists, otherwise throw an error.
`get_submodule`(target)	Return the submodule given by `target` if it exists, otherwise throw an error.
`half`()	Override default and prevent dtype casting.
`ipu`([device])	Move all model parameters and buffers to the IPU.
`load_state_dict`(state_dict[, strict, assign])	Copy parameters and buffers from `state_dict` into this module and its descendants.
`merge_state`(incoming_state)	Merge incoming metric state to the current state of the metric.
`modules`([remove_duplicate])	Return an iterator over all modules in the network.
`mtia`([device])	Move all model parameters and buffers to the MTIA.
`named_buffers`([prefix, recurse, ...])	Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.
`named_children`()	Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.
`named_modules`([memo, prefix, remove_duplicate])	Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.
`named_parameters`([prefix, recurse, ...])	Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.
`parameters`([recurse])	Return an iterator over module parameters.
`persistent`([mode])	Change post-init if metric states should be saved to its state_dict.
`plot`(_, *__)	Override this method plot the metric value.
`register_backward_hook`(hook)	Register a backward hook on the module.
`register_buffer`(name, tensor[, persistent])	Add a buffer to the module.
`register_forward_hook`(hook, *[, prepend, ...])	Register a forward hook on the module.
`register_forward_pre_hook`(hook, *[, ...])	Register a forward pre-hook on the module.
`register_full_backward_hook`(hook[, prepend])	Register a backward hook on the module.
`register_full_backward_pre_hook`(hook[, prepend])	Register a backward pre-hook on the module.
`register_load_state_dict_post_hook`(hook)	Register a post-hook to be run after module's `load_state_dict()` is called.
`register_load_state_dict_pre_hook`(hook)	Register a pre-hook to be run before module's `load_state_dict()` is called.
`register_module`(name, module)	Alias for `add_module()`.
`register_parameter`(name, param)	Add a parameter to the module.
`register_state_dict_post_hook`(hook)	Register a post-hook for the `state_dict()` method.
`register_state_dict_pre_hook`(hook)	Register a pre-hook for the `state_dict()` method.
`requires_grad_`([requires_grad])	Change if autograd should record operations on parameters in this module.
`reset`()	Promote the current epoch's min/max to be the bounds for the next epoch.
`set_dtype`(dst_type)	Transfer all metric state to specific dtype.
`set_extra_state`(state)	Set extra state contained in the loaded `state_dict`.
`set_submodule`(target, module[, strict])	Set the submodule given by `target` if it exists, otherwise throw an error.
`share_memory`()	See `torch.Tensor.share_memory_()`.
`state_dict`([destination, prefix, keep_vars])	Get the current state of metric as an dictionary.
`sync`([dist_sync_fn, process_group, ...])	Sync function for manually controlling when metrics states should be synced across processes.
`sync_context`([dist_sync_fn, process_group, ...])	Context manager to synchronize states.
`to`(args, *kwargs)
`to_empty`(*, device[, recurse])	Move the parameters and buffers to the specified device without copying storage.
`train`([mode])	Set the module in training mode.
`type`(dst_type)	Override default and prevent dtype casting.
`unsync`([should_unsync])	Unsync function for manually controlling when metrics states should be reverted back to their local states.
`update`(z)	Update the running min/max for the current epoch.
`xpu`([device])	Move all model parameters and buffers to the XPU.
`zero_grad`([set_to_none])	Reset gradients of all model parameters.

Attributes#

LatentStats.T_destination = ~T_destination#

LatentStats.call_super_init: bool = False#

LatentStats.device#: Return the device of the metric.

LatentStats.dtype#: Return the default dtype of the metric.

LatentStats.dump_patches: bool = False#

LatentStats.full_state_update: bool = False#

LatentStats.higher_is_better: bool | None = None#

LatentStats.is_differentiable: bool | None = None#

LatentStats.metric_state#: Get the current state of the metric.

LatentStats.plot_legend_name: str | None = None#

LatentStats.plot_lower_bound: float | None = None#

LatentStats.plot_upper_bound: float | None = None#

LatentStats.update_called#: Returns True if update or forward has been called initialization or last reset.

LatentStats.update_count#: Get the number of times update and/or forward has been called since initialization or last reset.

LatentStats.training: bool#

Methods#

LatentStats.add_module(name, module)#

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Return type:: None

Args:

name (str): name of the child module. The child module can be: accessed from this module using the given name

module (Module): child module to be added to the module.

LatentStats.add_state(name, default, dist_reduce_fx=None, persistent=False)#

Add metric state variable. Only used by subclasses.

Metric state variables are either :class:`~torch.Tensor or an empty list, which can be appended to by the metric. Each state variable must have a unique name associated with it. State variables are accessible as attributes of the metric i.e, if name is "my_state" then its value can be accessed from an instance metric as metric.my_state. Metric states behave like buffers and parameters of Module as they are also updated when .to() is called. Unlike parameters and buffers, metric states are not by default saved in the modules state_dict.

Return type:: None

Args:

name: The name of the state variable. The variable will then be accessible at self.name. default: Default value of the state; can either be a Tensor or an empty list.

The state will be reset to this value when self.reset() is called.

dist_reduce_fx (Optional): Function to reduce state across multiple processes in distributed mode.: If value is "sum", "mean", "cat", "min" or "max" we will use torch.sum, torch.mean, torch.cat, torch.min and torch.max` respectively, each with argument dim=0. Note that the "cat" reduction only makes sense if the state is a list, and not a tensor. The user can also pass a custom function in this parameter.
persistent (Optional): whether the state will be saved as part of the modules state_dict.: Default is False.

Note

Setting dist_reduce_fx to None will return the metric state synchronized across different processes. However, there won’t be any reduction function applied to the synchronized metric state.

The metric states would be synced as follows

If the metric state is Tensor, the synced value will be a stacked Tensor across the process dimension if the metric state was a Tensor. The original Tensor metric state retains dimension and hence the synchronized output will be of shape (num_process, ...).
If the metric state is a list, the synced value will be a list containing the combined elements from all processes.

Important

When passing a custom function to dist_reduce_fx, expect the synchronized metric state to follow the format discussed in the above note.

Caution

The values inserted into a list state are deleted whenever reset() is called. This allows device memory to be automatically reallocated, but may produce unexpected effects when referencing list states. To retain such values after reset() is called, you must first copy them to another object.

Raises:

ValueError:: If default is not a tensor or an empty list.
ValueError:: If dist_reduce_fx is not callable or one of "mean", "sum", "cat", "min", "max" or None.

LatentStats.apply(fn)#

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also torch.nn.init).

Return type:: Self

Args:: fn (Module -> None): function to be applied to each submodule
Returns:: Module: self

Example:

>>> @torch.no_grad()
>>> def init_weights(m):
>>>     print(m)
>>>     if type(m) is nn.Linear:
>>>         m.weight.fill_(1.0)
>>>         print(m.weight)
>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

LatentStats.bfloat16()#

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns:: Module: self

Return type:: Self

LatentStats.buffers(recurse=True)#

Return an iterator over module buffers.

Return type:: Iterator[Tensor]

Args:

recurse (bool): if True, then yields buffers of this module: and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields:

torch.Tensor: module buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for buf in model.buffers():
>>>     print(type(buf), buf.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

LatentStats.children()#

Return an iterator over immediate children modules.

Return type:: Iterator[Module]

Yields:: Module: a child module

LatentStats.clone()#

Make a copy of the metric.

Return type:: Metric

LatentStats.compile(*args, **kwargs)#

Compile this Module’s forward using torch.compile().

This Module’s __call__ method is compiled and all arguments are passed as-is to torch.compile().

See torch.compile() for details on the arguments for this function.

Return type:: None

LatentStats.compute()[source]#: Compute vanished statistics based on the previous epoch’s bounds.

LatentStats.cpu()#

Move all model parameters and buffers to the CPU.

Note

This method modifies the module in-place.

Returns:: Module: self

Return type:: Self

LatentStats.cuda(device=None)#

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

Note

This method modifies the module in-place.

Args:

device (int, optional): if specified, all parameters will be: copied to that device

Returns:

Module: self

Return type:: Self

LatentStats.double()#

Override default and prevent dtype casting.

Please use Metric.set_dtype() instead.

Return type:: Metric

LatentStats.eval()#

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False).

See Locally disabling gradient computation for a comparison between .eval() and several similar mechanisms that may be confused with it.

Return type:: Self

Returns:: Module: self

LatentStats.extra_repr()#

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

Return type:: str

LatentStats.float()#

Override default and prevent dtype casting.

Please use Metric.set_dtype() instead.

Return type:: Metric

LatentStats.forward(*args, **kwargs)#

Aggregate and evaluate batch input directly.

Serves the dual purpose of both computing the metric on the current batch of inputs but also add the batch statistics to the overall accumulating metric state. Input arguments are the exact same as corresponding update method. The returned output is the exact same as the output of compute.

Return type:: Any

Args:

args: Any arguments as required by the metric update method. kwargs: Any keyword arguments as required by the metric update method.

Returns:

The output of the compute method evaluated on the current batch.

Raises:

TorchMetricsUserError:: If the metric is already synced and forward is called again.

LatentStats.get_buffer(target)#

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Return type:: Tensor

Args:

target: The fully-qualified string name of the buffer: to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns:

torch.Tensor: The buffer referenced by target

Raises:

AttributeError: If the target string references an invalid: path or resolves to something that is not a buffer

LatentStats.get_extra_state()#

Return any extra state to include in the module’s state_dict.

Implement this and a corresponding set_extra_state() for your module if you need to store extra state. This function is called when building the module’s state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Return type:: Any

Returns:: object: Any extra state to store in the module’s state_dict

LatentStats.get_latest_min_max()[source]#

LatentStats.get_parameter(target)#

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Return type:: Parameter

Args:

target: The fully-qualified string name of the Parameter: to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns:

torch.nn.Parameter: The Parameter referenced by target

Raises:

AttributeError: If the target string references an invalid: path or resolves to something that is not an nn.Parameter

LatentStats.get_submodule(target)#

Return the submodule given by target if it exists, otherwise throw an error.

For example, let’s say you have an nn.Module A that looks like this:

A(
    (net_b): Module(
        (net_c): Module(
            (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2))
        )
        (linear): Linear(in_features=100, out_features=200, bias=True)
    )
)

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Return type:: Module

Args:

target: The fully-qualified string name of the submodule: to look for. (See above example for how to specify a fully-qualified string.)

Returns:

torch.nn.Module: The submodule referenced by target

Raises:

AttributeError: If at any point along the path resulting from: the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

LatentStats.half()#

Override default and prevent dtype casting.

Please use Metric.set_dtype() instead.

Return type:: Metric

LatentStats.ipu(device=None)#

Move all model parameters and buffers to the IPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on IPU while being optimized.

Note

This method modifies the module in-place.

Arguments:

device (int, optional): if specified, all parameters will be: copied to that device

Returns:

Module: self

Return type:: Self

LatentStats.load_state_dict(state_dict, strict=True, assign=False)#

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Warning

If assign is True the optimizer must be created after the call to load_state_dict unless get_swap_module_params_on_conversion() is True.

Args:

state_dict (dict): a dict containing parameters and: persistent buffers.
strict (bool, optional): whether to strictly enforce that the keys: in state_dict match the keys returned by this module’s state_dict() function. Default: True
assign (bool, optional): When set to False, the properties of the tensors: in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of Parameter for which the value from the module is preserved. Default: False

Returns:

NamedTuple with missing_keys and unexpected_keys fields:

missing_keys is a list of str containing any keys that are expected
by this module but missing from the provided state_dict.
unexpected_keys is a list of str containing the keys that are not
expected by this module but present in the provided state_dict.

Note:

If a parameter or buffer is registered as None and its corresponding key exists in state_dict, load_state_dict() will raise a RuntimeError.

LatentStats.merge_state(incoming_state)#

Merge incoming metric state to the current state of the metric.

Return type:: None

Args:

incoming_state:: either a dict containing a metric state similar to the metric itself or an instance of the metric class.

Raises:

ValueError:: If the incoming state is neither a dict nor an instance of the metric class.
RuntimeError:: If the metric has full_state_update=True or dist_sync_on_step=True. In these cases, the metric cannot be merged with another metric state in a simple way. The user should overwrite the method in the metric class to handle the merge operation.
ValueError:: If the incoming state is a metric instance but the class is different from the current metric class.

Example with a metric instance:

>>> from torchmetrics.aggregation import SumMetric
>>> metric1 = SumMetric()
>>> metric2 = SumMetric()
>>> metric1.update(1)
>>> metric2.update(2)
>>> metric1.merge_state(metric2)
>>> metric1.compute()
tensor(3.)

Example with a dict:

>>> from torchmetrics.aggregation import SumMetric
>>> metric = SumMetric()
>>> metric.update(1)
>>> # SumMetric has one state variable called `sum_value`
>>> metric.merge_state({"sum_value": torch.tensor(2)})
>>> metric.compute()
tensor(3.)

LatentStats.modules(remove_duplicate=True)#

Return an iterator over all modules in the network.

Return type:: Iterator[Module]

Args:

remove_duplicate: whether to remove the duplicated module instances in the result: or not.

Yields:

Module: a module in the network

Note:

Duplicate modules are returned only once by default. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.modules()):
...     print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

LatentStats.mtia(device=None)#

Move all model parameters and buffers to the MTIA.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on MTIA while being optimized.

Note

This method modifies the module in-place.

Arguments:

device (int, optional): if specified, all parameters will be: copied to that device

Returns:

Module: self

Return type:: Self

LatentStats.named_buffers(prefix='', recurse=True, remove_duplicate=True)#

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Return type:: Iterator[tuple[str, Tensor]]

Args:

prefix (str): prefix to prepend to all buffer names. recurse (bool, optional): if True, then yields buffers of this module

and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.

remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields:

(str, torch.Tensor): Tuple containing the name and buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, buf in self.named_buffers():
>>>     if name in ['running_var']:
>>>         print(buf.size())

LatentStats.named_children()#

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Return type:: Iterator[tuple[str, Module]]

Yields:: (str, Module): Tuple containing a name and child module

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, module in model.named_children():
>>>     if name in ['conv4', 'conv5']:
>>>         print(module)

LatentStats.named_modules(memo=None, prefix='', remove_duplicate=True)#

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args:: memo: a memo to store the set of modules already added to the result prefix: a prefix that will be added to the name of the module remove_duplicate: whether to remove the duplicated module instances in the result

or not
Yields:: (str, Module): Tuple of name and module
Note:: Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.named_modules()):
...     print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

LatentStats.named_parameters(prefix='', recurse=True, remove_duplicate=True)#

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Return type:: Iterator[tuple[str, Parameter]]

Args:

prefix (str): prefix to prepend to all parameter names. recurse (bool): if True, then yields parameters of this module

and all submodules. Otherwise, yields only parameters that are direct members of this module.

remove_duplicate (bool, optional): whether to remove the duplicated: parameters in the result. Defaults to True.

Yields:

(str, Parameter): Tuple containing the name and parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, param in self.named_parameters():
>>>     if name in ['bias']:
>>>         print(param.size())

LatentStats.parameters(recurse=True)#

Return an iterator over module parameters.

This is typically passed to an optimizer.

Return type:: Iterator[Parameter]

Args:

recurse (bool): if True, then yields parameters of this module: and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields:

Parameter: module parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for param in model.parameters():
>>>     print(type(param), param.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

LatentStats.persistent(mode=False)#

Change post-init if metric states should be saved to its state_dict.

Return type:: None

LatentStats.plot(*_, **__)#

Override this method plot the metric value.

Return type:: Any

LatentStats.register_backward_hook(hook)#

This function is deprecated in favor of register_full_backward_hook() and the behavior of this function will change in future versions.

Return type:: RemovableHandle

Returns:

torch.utils.hooks.RemovableHandle:: a handle that can be used to remove the added hook by calling handle.remove()

LatentStats.register_buffer(name, tensor, persistent=True)#

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the module’s state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module’s state_dict.

Buffers can be accessed as attributes using given names.

Return type:: None

Args:

name (str): name of the buffer. The buffer can be accessed: from this module using the given name
tensor (Tensor or None): buffer to be registered. If None, then operations: that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module’s state_dict.
persistent (bool): whether the buffer is part of this module’s: state_dict.

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> self.register_buffer('running_mean', torch.zeros(num_features))

LatentStats.register_forward_hook(hook, *, prepend=False, with_kwargs=False, always_call=False)#

The hook will be called every time after forward() has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward() is called. The hook should have the following signature:

hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature:

hook(module, args, kwargs, output) -> None or modified output

Return type:: RemovableHandle

Args:

hook (Callable): The user defined hook to be registered. prepend (bool): If True, the provided hook will be fired

before all existing forward hooks on this torch.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this torch.nn.Module. Note that global forward hooks registered with register_module_forward_hook() will fire before all hooks registered by this method. Default: False

with_kwargs (bool): If True, the hook will be passed the: kwargs given to the forward function. Default: False
always_call (bool): If True the hook will be run regardless of: whether an exception is raised while calling the Module. Default: False

Returns:

torch.utils.hooks.RemovableHandle:: a handle that can be used to remove the added hook by calling handle.remove()

LatentStats.register_forward_pre_hook(hook, *, prepend=False, with_kwargs=False)#

The hook will be called every time before forward() is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature:

hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature:

hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Return type:: RemovableHandle

Args:

hook (Callable): The user defined hook to be registered. prepend (bool): If true, the provided hook will be fired before

all existing forward_pre hooks on this torch.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this torch.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook() will fire before all hooks registered by this method. Default: False

with_kwargs (bool): If true, the hook will be passed the kwargs: given to the forward function. Default: False

Returns:

torch.utils.hooks.RemovableHandle:: a handle that can be used to remove the added hook by calling handle.remove()

LatentStats.register_full_backward_hook(hook, prepend=False)#

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.

If none of the module inputs require gradients, the hook will fire when the gradients are computed with respect to module outputs.

If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature:

hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module’s forward function.

Warning

Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error.

Args:

hook (Callable): The user-defined hook to be registered. prepend (bool): If true, the provided hook will be fired before

all existing backward hooks on this torch.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this torch.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook() will fire before all hooks registered by this method.

Returns:

torch.utils.hooks.RemovableHandle:: a handle that can be used to remove the added hook by calling handle.remove()

Return type:: RemovableHandle

LatentStats.register_full_backward_pre_hook(hook, prepend=False)#

The hook will be called every time the gradients for the module are computed. The hook should have the following signature:

hook(module, grad_output) -> tuple[Tensor, ...], Tensor or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

Warning

Modifying inputs inplace is not allowed when using backward hooks and will raise an error.

Args:

hook (Callable): The user-defined hook to be registered. prepend (bool): If true, the provided hook will be fired before

all existing backward_pre hooks on this torch.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this torch.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook() will fire before all hooks registered by this method.

Returns:

torch.utils.hooks.RemovableHandle:: a handle that can be used to remove the added hook by calling handle.remove()

Return type:: RemovableHandle

LatentStats.register_load_state_dict_post_hook(hook)#

It should have the following signature::: hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict() with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns:

torch.utils.hooks.RemovableHandle:: a handle that can be used to remove the added hook by calling handle.remove()

LatentStats.register_load_state_dict_pre_hook(hook)#

It should have the following signature::

hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None # noqa: B950

Arguments:

hook (Callable): Callable hook that will be invoked before: loading the state dict.

LatentStats.register_module(name, module)#

Alias for add_module().

Return type:: None

LatentStats.register_parameter(name, param)#

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Return type:: None

Args:

name (str): name of the parameter. The parameter can be accessed: from this module using the given name
param (Parameter or None): parameter to be added to the module. If: None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module’s state_dict.

LatentStats.register_state_dict_post_hook(hook)#

It should have the following signature::: hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

LatentStats.register_state_dict_pre_hook(hook)#

It should have the following signature::: hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

LatentStats.requires_grad_(requires_grad=True)#

Change if autograd should record operations on parameters in this module.

This method sets the parameters’ requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See Locally disabling gradient computation for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Return type:: Self

Args:

requires_grad (bool): whether autograd should record operations on: parameters in this module. Default: True.

Returns:

Module: self

LatentStats.reset()[source]#: Promote the current epoch’s min/max to be the bounds for the next epoch.

LatentStats.set_dtype(dst_type)#

Transfer all metric state to specific dtype. Special version of standard type method.

Return type:: Metric

Arguments:: dst_type: the desired type as string or dtype object

LatentStats.set_extra_state(state)#

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Return type:: None

Args:: state (dict): Extra state from the state_dict

LatentStats.set_submodule(target, module, strict=False)#

Set the submodule given by target if it exists, otherwise throw an error.

Note

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let’s say you have an nn.Module A that looks like this:

A(
    (net_b): Module(
        (net_c): Module(
            (conv): Conv2d(3, 3, 3)
        )
        (linear): Linear(3, 3)
    )
)

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args:

target: The fully-qualified string name of the submodule: to look for. (See above example for how to specify a fully-qualified string.)

module: The module to set the submodule to. strict: If False, the method will replace an existing submodule

or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn’t already exist.

Raises:

ValueError: If the target string is empty or if module is not an instance of nn.Module. AttributeError: If at any point along the path resulting from

the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

Return type:: None

LatentStats.share_memory()#

See torch.Tensor.share_memory_().

Return type:: Self

LatentStats.state_dict(destination=None, prefix='', keep_vars=False)#

Get the current state of metric as an dictionary.

Return type:: dict[str, Any]

Args:

destination: Optional dictionary, that if provided, the state of module will be updated into the dict and: the same object is returned. Otherwise, an OrderedDict will be created and returned.

prefix: optional string, a prefix added to parameter and buffer names to compose the keys in state_dict. keep_vars: by default the Tensor returned in the state dict are detached from autograd.

If set to True, detaching will not be performed.

LatentStats.sync(dist_sync_fn=None, process_group=None, should_sync=True, distributed_available=None)#

Sync function for manually controlling when metrics states should be synced across processes.

Return type:: None

Args:

dist_sync_fn: Function to be used to perform states synchronization process_group:

Specify the process group on which synchronization is called. default: None (which selects the entire world)

should_sync: Whether to apply to state synchronization. This will have an impact: only when running in a distributed setting.

distributed_available: Function to determine if we are running inside a distributed setting

Raises:

TorchMetricsUserError:: If the metric is already synced and sync is called again.

LatentStats.sync_context(dist_sync_fn=None, process_group=None, should_sync=True, should_unsync=True, distributed_available=None)#

Context manager to synchronize states.

This context manager is used in distributed setting and makes sure that the local cache states are restored after yielding the synchronized state.

Return type:: Generator

Args:

dist_sync_fn: Function to be used to perform states synchronization process_group:

Specify the process group on which synchronization is called. default: None (which selects the entire world)

should_sync: Whether to apply to state synchronization. This will have an impact: only when running in a distributed setting.
should_unsync: Whether to restore the cache state so that the metrics can: continue to be accumulated.

distributed_available: Function to determine if we are running inside a distributed setting

LatentStats.to(*args, **kwargs)#

Overloads:

self, device (DeviceLikeType | None), dtype (dtype | None), non_blocking (bool) → Self
self, dtype (dtype), non_blocking (bool) → Self
self, tensor (Tensor), non_blocking (bool) → Self

Move and/or cast the parameters and buffers.

This can be called as

to(device=None, dtype=None, non_blocking=False)

drvi.internal.drvi.to(dtype, non_blocking=False)

drvi.internal.drvi.to(tensor, non_blocking=False)

drvi.internal.drvi.to(memory_format=torch.channels_last)

Its signature is similar to torch.Tensor.to(), but only accepts floating point or complex dtypes. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

Note

This method modifies the module in-place.

Args:

device (torch.device): the desired device of the parameters: and buffers in this module
dtype (torch.dtype): the desired floating point or complex dtype of: the parameters and buffers in this module
tensor (torch.Tensor): Tensor whose dtype and device are the desired: dtype and device for all parameters and buffers in this module
memory_format (torch.memory_format): the desired memory: format for 4D parameters and buffers in this module (keyword only argument)

Returns:

Module: self

Examples:

>>> # xdoctest: +IGNORE_WANT("non-deterministic")
>>> linear = nn.Linear(2, 2)
>>> linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
>>> linear.to(torch.double)
Linear(in_features=2, out_features=2, bias=True)
>>> linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=torch.float64)
>>> # xdoctest: +REQUIRES(env:TORCH_DOCTEST_CUDA1)
>>> gpu1 = torch.device("cuda:1")
>>> linear.to(gpu1, dtype=torch.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
>>> linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=torch.float16, device='cuda:1')
>>> cpu = torch.device("cpu")
>>> linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
>>> linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=torch.float16)

>>> linear = nn.Linear(2, 2, bias=None).to(torch.cdouble)
>>> linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=torch.complex128)
>>> linear(torch.ones(3, 2, dtype=torch.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=torch.complex128)

LatentStats.to_empty(*, device, recurse=True)#

Move the parameters and buffers to the specified device without copying storage.

Return type:: Self

Args:

device (torch.device): The desired device of the parameters: and buffers in this module.
recurse (bool): Whether parameters and buffers of submodules should: be recursively moved to the specified device.

Returns:

Module: self

LatentStats.train(mode=True)#

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Return type:: Self

Args:

mode (bool): whether to set training mode (True) or evaluation: mode (False). Default: True.

Returns:

Module: self

LatentStats.type(dst_type)#

Override default and prevent dtype casting.

Please use Metric.set_dtype() instead.

Return type:: Metric

LatentStats.unsync(should_unsync=True)#

Unsync function for manually controlling when metrics states should be reverted back to their local states.

Return type:: None

Args:: should_unsync: Whether to perform unsync

LatentStats.update(z)[source]#: Update the running min/max for the current epoch.

LatentStats.xpu(device=None)#

Move all model parameters and buffers to the XPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized.

Note

This method modifies the module in-place.

Arguments:

device (int, optional): if specified, all parameters will be: copied to that device

Returns:

Module: self

Return type:: Self

LatentStats.zero_grad(set_to_none=True)#

Reset gradients of all model parameters.

See similar function under torch.optim.Optimizer for more context.

Return type:: None

Args:

set_to_none (bool): instead of setting to zero, set the grads to None.: See torch.optim.Optimizer.zero_grad() for details.

drvi.internal.drvi.LatentStats

Contents

drvi.internal.drvi.LatentStats#

Attributes table#

Methods table#

Attributes#

Methods#